It will come as news to none that the Epstein Files have been (partially) released despite being HEAVILY redacted. This is against the law. The laws of the United Stated are enforced by the Executive Branch, headed by the President. Currently, it is VERY CLEAR the Department of Justice has violated the laws of the United States with how they have released this files.

But that’s not important right now (well, it is, but they are doing so in many, many, many ways, this is but one of them and not the *main* point here). This is about what happens when justice is corrupted beyond recognition.

TLDR; (I have written scripts to download all the files (repo link), been doing so, found some wild stuff, and will share all of it below.)

UPDATE: This is a great interface for looking through the files! (although the back functionality is still wonked, be careful)

UPDATE2: Here’s the raw files in a nice folder I don’t have to pay for. I’ll have to check if I have any they are missing.

Executive Summary: This is a bit long, but this post is mostly about a number of very powerful and rich men, people like Donald Trump (current President of the United States), and billionaires named Leon Black and Marc Rowan, the founders of Apollo Global Management (which is currently backing the hostile bid for Warner Bros by Paramount). This is about a wide-ranging decades long conspiracy involving the abuse of thousands of girls and what appears to be an international blackmail operation used to get inside information on…everything. It has made the men involved supremely wealthy. The subtext of this entire piece is extreme wealth inequity and what kind of world that creates.

Leon Black (who as the SEC documented, gave Epstein over $160,000,000 *after* his 2008 conviction):

Background: The Epstein Media

I put together the video above prior to the election in 2024 so people would know what was at stake (from the Epstein angle). It got no traction, but whatever, hindsight continues its long 20/20 streak and so let’s dive in.***

[Hold on…I realized there’s one more thing we need to make sure everyone knows. THIS WAS NOT A SECRET. Not at all. In files is a copy of this NY Mag article from 2006, “Billionares Are Free”. In the files, the victim’s name is redacted. In the public article, it is not. That article lays it all out (Haley Robson has since publicly testified about this many times).

The next paragraph lays out the moral outrage, at the time…2006 (20 years ago, not a typo).

So, it was an open secret. What was happening behind the headlines? LFG!!

First up. The Files. You can try and see them here: The Epstein Library at the Department of Justice.

Some things to note about that page. The search function is only for plain text. This means that documents consisting of handwritten notes cannot be found via the search. The same is true for PDF’s of photographs. Also, you must authenticate every five minutes. So after five minutes, you will get the age verification and human verification pop-ups again. This makes automated downloading of the files problematic. I didn’t totally solve it, but check the repo for what we are going for, I’ll need some help for that last step (automated re-authentication), I think. The scripts are all here (python).**

The numbers next to each Data Set release below correspond to the document numbers printed on each artifact. However, you’ll see skips in this ordering in the listed files below in the DOJ library. This is because multiple page PDF’s receive a number *for each page*. But they are only listed (and therefore accessible) by the first document number in the set. So for example, this copy of “Massage for Dummies” is document numbers 8020 – 8119, but is only accessible via Document 8020 (same pattern holds for the rest of the book).

They have listed all the files available is each release here (click “Epstein Files Transparency Act (H.R4405” to list them).

Data Set 1 (1 – 3,158) [Epstein’s home in Manhattan, the largest personal residence in the borough. As you ascend the staircase to the top floor (where the massage tables are), you are literally treated like you are in Heaven, above the clouds. Included in this set are lots of pictures of evidence (picture albums, hard drives) collected from his residence. These show up more in Data Set 3. Looking through his home is surreal. You’ll see the very inconsistent redactions (like paintings and adult magazine covers) that should raise alarms (i.e. over-redactions done by an AI/machine, with lots of false positives). This is the home mentioned in the NY Mag article quoted above.

Data Set 2 (3,159 – 3,857) [More personal vacation photos, celebrities, lots of photos. These folks documented *everything*. This is where we start to see the full page retractions which usually indicate Trump pictures (3236 for example).] You can also see the pattern when there are victims, they only the faces. But with Donald, it’s the full photo. (yes, that’s Mick Jagger). NOTE: this is where we start to see fully missing files, more on that below.

Data Set 3 (3,858 – 5,704) [This is where we start to get into the contents of the stuff photographed in Data Set 1. Those include all the photo albums and we see pictures of many of the hard drives. We are also start see many of the images like the one below “CSAM NOT SCANNED”. There are LOTS of these and all are evidence of crimes. Child sex abuse crimes. All of them. And there are a LOT of these. Again, the number of actual victims is in the 1,000’s. They had access to the entire world for decades. The ones that have been able to come forward are a tiny slice of the whole pool. 4348 likely included pictures of Donald prior to redactions (see documents 4362-4366 for examples). 5386 is another one where victims are partially redacted but some photos are fully redacted.**** A lot of these are PDF’s of pictures taken of physical photo albums. Quality is often degraded simply because of this process.

Data Set 4 (5,705 – 8,408) [Internal FBI docs and docs of evidence. (like the Idiot’s Guide to Massage). Some very suspicious redactions (5848, for example).

Data Set 5 (8,409 – 8,584) [Smaller release, mostly pictures of digital evidence]

Data Set 6 (8,585 – 9,015) [Legal documents and filings for Epstein and Maxwell. Including the grand jury docs and formal presentations. No missing files]

Data Set 7 (9,016 – 9,675) [Internal FBI docs and interviews. Oddly, the first document 9016 starts at page 200 of an interview. This means the first 200 pages are not available. That’s some missing docs. These also include some handwritten stuff and victim interviews. (see: “9th Grade”)

Data Set 8 (9,676 – 39,025) [Internal FBI stuff…lots of victim testimony and tips…very rough stuff…with many redactions. I have not, obviously, been able to review all these docs. Up to now I think I have at least glanced at everything. This has been disturbing and after posting will be taking a break. For example: Document 20517 . Read that while knowing (with hindsight)…

A 19-year-old student from Kiefer died on Jan. 10, 2000, according to death notices in two newspapers, the Tulsa World and the Sapulpa Herald. The state’s chief medical examiner, Dr. Eric Pfeifer, told The Oklahoman on Dec. 24 that the death of Dusti Rhea Duke was ruled a suicide by a gunshot to the head.

…if the FBI would unredact that one, we’d know how accurate this tip is. Considering what else is in it…

“a few seconds from pulling the limousine over on the
median and within a few seconds of pulling him out of the car and hurting
him due to some of things he was saying” as caller choose not to. III
noted Trump continuously stated the name “Jeffrey” while on the phone,
and made references to “abusing some girl.” III was unsure who he was
talking to nor who he was referencing. As III talked about his time
meeting Donald Trump, immediately demeanor went “stone
cold” as stated “he raped me.” III said “what” as replied
“Donald J. Trump had raped her along with Jeffrey Epstein.” noted
some girl with a funny name “took me into a fancy hotel or building,
that’s how it happened.”

II advised to call the police regarding the incident as
stated “I can’t they will kill me.” On Christmas Day, contacted III
stated she had in fact called the police about what they had talked
about. III told she had “done good.” III did not hear from or
until 01/10/2000.

UPDATE: Data Set 8 also includes lots of video footage from Epstein’s cell and various documents around his murder/suicide. These are clearly labeled as mp4’s (video files) in the DOJ’s site. However, because they have the same document id as a PDF, they can be missed. I found that by looking at the size of the PDF (3kb) and reading it (“No Images Produced”), we can quickly also grab any mp4 at the same URL. So now the scripts automatically grab mp4’s. There is another pattern I’ve noticed but haven’t…(aaah!)…if the file is 5kb and has the words “Native Placeholder” they appear to be Excel files, m4a (audio) files, or CSV’s (raw text files…quite redacted, it turns out). As you can access here. The scripts now automatically handle each known scenario. (I think this will matter for DS 9-11, where we have millions of files to filter).

Data Set 9 (39,025 – 1,262,781) [We are now in the internal FBI emails and massive amounts of irrelevant docs. Almost all internal FBI folks have emails redacted. This makes traversing conversations difficult. Lots of unanswered questions are raised. Lots of curious docs, I will explore these in another post. This is also where a lot of personal emails show up. I would highly recommend to use the JMAIL archive to explore these. It is a wonderful tool to explore the archive. But you have to search to find stuff like this in 474365 (which names a BUNCH of people he’s meeting with in 2018).] Note: You can also find PDF’s of unredacted CSV’s, like this one of Epstein’s (?) flight logs. This is an artifact of automated conversions. I’ve found a few places where the formal doc was redacted, but the PDF of the raw CSV was not.

RED FLAG NOTE: NO DOCUMENTS ABOVE number 418425 are available via the DOJ’s manual web interface. It stops at page 1957, and no pages are available beyond that. That’s nearly 800,000 documents that are…only available via search (?). Looking into this (2/15)….I have confirmed you can search for stuff like this *and* you can download it directly, BUT you can’t find it directly via the web interface (the “pages” stop at 1957).

Data Set 10 (1,262,782 – 2.217,416) [Note: how we are able to get a page over 9,000 here (10,027 to be exact) from the DOJ’s website. This means the limitation noticed in Data Set 9 is *not* a publishing software issue. There is also a crossover with Data Set 11 for the numbering, with about 12,000 crossover numbers (i.e. the same numbered document is in both places…digging in 2/15…)]

Data Set 11 (2,205,655 – 2,730,264) [Starts with a big missing file (s). First Doc ID is 2,205,655 through 2,205,669. The next Doc ID is 2,205,827. So Documents 2,205,670 – 2,205,826 are MISSING! Note: There may be a good explanation for this, but given the pattern and the totality of the law requiring the release itself, all missing files have to have the Attorney General explain *why*. These aren’t “redactions” these are straight up missing files.

This pattern continues, as files 2,205,843 – 2,209,621 are also missing. That’s almost four thousand missing files to *start* the Dataset. This pattern continues for a while. Coming back to this later, it’s time to publish.

Data Set 12 (2,730,265 – 2,731,852) [ This has some concerning redactions, both literally and in the abstract sense of “justice itself”.]

Once I realized how they are ordering these (perfectly sequential as the law requires ALL files to be exposed), you realize there are missing files. Many missing files. See above for some examples, but there are many more. I will update on a full post later.

Missing files leave holes, holes can be explored for depth, width, and proximity.

Ok, that’s it. A quick walk through the 12 DataSets. As of February 16th, 2026, the Department of Justice of the United States is claiming that all relevant files have been publicly released. This appears to be a bald-faced lie (as this document documents and you can do for yourself with the code provided below).

These are dangerous times to know things.

*fin*

** About the Repo: The scripts are very simple. These are Python scripts. You should be able to clone the repo to your local machine, install the requirements and run the scripts. Instructions are in the readme for command-line switches for authentication and “fast as possible” downloads. Authentication only lasts for five minutes, as alluded to above. I had originally put in a delay between file downloads, but that doesn’t appear to be necessary (hence the “–no-pause” switch option) . So I’ve been working on Datasets 8-11 for a while (ongoing as we push to publish).

Also, it has been revealed that other files types (outside of PDF) have also been posted. These are often listed directly in the web interface on the DOJ’s website if you make to the page they are on. Other than that, you get the PDF pattern we discussed up in Data Set 8. I had attempted something earlier to get the other file types, but missed the correct pattern until getting a tip from a friend.

I suspected this for the missing files and adding some stuff in the scripts to look for other files types for missing files. I have since learned that the videos are available as “.avi” or “.mp4” files and HAVE THE SAME DOCUMENT ID AS OTHER PDF’s! This was not one of my assumptions (I had logically excluded it). The DOJ has since removed some of these files, which is why I think this effort is worthwhile, even now. There’s more stuff here. My assumption is they are going with “security through obscurity” and are not confident they have scrubbed all of Donald’s presence, but *also* the vast majority of them (DOJ) want to “technically” follow the law, so the people actually doing the work (DOJ) are likely to have included stuff their bosses bosses boss wouldn’t want in there.

I have now been able to identify the pattern and the scripts download “.mp4” (video), “.m4a” (audio), “.csv” (plain text), and “.xlxs” (Excel) files when it finds a particular size PDF with particular text.

Update: The script is now working for “.mp3” (audio), “.mov” (video), “.wav” (audio) files. This has been applied to Scripts for Data Set 8-11 (when I push).

Yes, Those Files code repository.

*** “Hindsight is 20/20” is a phrase I used often, as someone who has suffered from poor vision since youth. The future is unwritten, and the past cannot be changed, everything new actually happens right now. One can spend the now studying the past, or working to define the future, but not both at the same time. This work is an effort to get a 20/20 look at the past, so that we may better define a future where our culture isn’t owned and controlled by a bunch of ruthless child rapists.

****As of 2/14/26 Document 5386 is no longer publicly available via the DOJ’s website. It contains highly sensitive stuff, likely including Donald (redacted). I am not sure how many more documents I was able to get before they realized what they were, but there are more. Many more, IMVIHO. I would suggest anyone doing this kind of work to do a “diff” against what is available now vs what they may have been able to download previously and start with those missing docs for deeper review/FOIA targets.

***** I am publishing this on President’s Day 2026. As you can see from this post, at least two United Stated Presidents have been closely involved with a conspiracy to cover up a group of billionaire’s who raped children over the course of multiple decades. It is time for a reckoning for DONALD TRUMP. His department of Justice has attempted to *METICUOUSLY ERASE* his name and image from these files. But they have failed. Every shadow you see, every black space in this evil endeavor is filled with his face and his presence.

******I have very little doubt that Epstein did not kill himself. I have a decent idea (I think, until I see exculpatory evidence) of who killed him.

*******As anyone who has read these files and understand who exactly people like Donald Trump, Leon Black, and Marc Rowan are and are capable of, I must state plainly, I am not suicidal. I live (quite frankly) a life that looks like the American Dream, with a beautiful, intelligent, capable, accomplished spouse, four incredible children, and all the stresses and joys that come with suburban life in the 21st century. There is a reason everyone who deeply reads and understands this *conspiracy* says stuff like this. It didn’t get this far or go this long, without violence.

Happy Presidents’ Day.